Light emitting diodes are known which, when disposed on a circuit, accept electrical impulses from the circuit and convert the impulses into light signals. LEDs are energy efficient, they give off virtually no heat, and they have a long lifetime.
A number of types of LED exist, including air gap LEDs, GaAs light-emitting diodes (which may be doubled and packaged as single unit offer greater reliability than conventional single-diode package), polymer LEDs, and semi-conductor LEDs, among others. Most LEDs in current use are red. Conventional uses for LEDs include displays for low light environments, such as the flashing light on a modem or other computer component, or the digital display of a wristwatch. Improved LEDs have recently been used in arrays for longer-lasting traffic lights. LEDs have been used in scoreboards and other displays. Also, LEDs have been placed in arrays and used as television displays. Although most LEDs in use are red, yellow or white, LEDs may take any color; moreover, a single LED may be designed to change colors to any color in the color spectrum in response to changing electrical signals.
It is well known that combining the projected light of one color with the projected light of another color will result in the creation of a third color. It is also well known that three commonly used primary colors--red, blue and green--can be combined in different proportions to generate almost any color in the visible spectrum. The present invention takes advantage of these effects by combining the projected light from at least two light emitting diodes (LEDS) of different primary colors. It should be understood that for purposes of this invention the term "primary colors" encompasses any different colors that can be combined to create other colors.
Computer lighting networks that use LEDs are also known. U.S. Pat. No. 5,420,482, issued to Phares, describes one such network that uses different colored LEDs to generate a selectable color, primarily for use in a display apparatus. U.S. Pat. No. 4,845,481, issued to Havel, is directed to a multicolored display device. Havel uses a pulse width modulated signal to provide current to respective LEDs at a particular duty cycle. U.S. Pat. No. 5,184,114, issued to Brown, shows an LED display system. U.S. Pat. No. 5,134,387, issued to Smith et al., is directed to an LED matrix display.
Illumination systems exist in which a network of individual lights is controlled by a central driver, which may be a computer-controlled driver. Such illumination systems include theatrical lighting systems. The USITT DMX-512 protocol was developed to deliver a stream of data from a theatrical console to a series of theatrical lights.
The DMX-512 protocol was originally designed to standardize the control of light dimmers by lighting consoles. The DMX-512 protocol is a multiplexed digital lighting control protocol with a signal to control 512 devices, such device including dimmers, scrollers, non-dim relays, parameters of a moving light, or a graphical light in a computerized virtual reality set. DMX-512 is used for control for a network of devices. The DMX-512 protocol employs digital signal codes. When a transmitting device, such as a lighting console, sends digital codes, a receiving device, such as a dimmer, transforms these codes into a function command, such as dimming to a specified level. With digital systems, signal integrity is compromised less over long cable runs, relative to analog control. When a coded string of 0/1 digits are sent and received, the device will perform the desired task.
In hardware terms, DMX-512 protocol information is transferred between devices over metal wires using the RS-485 hardware protocol. This involves the use of two wires, known as a twisted pair. The first wire is referred to as a data+wire, and the second wire is referred to as a data-wire. The voltage used on the line is typically positive five volts. By way of example, to transmit a logical one, the data+wire is taken to positive five volts, and the data-wire to zero volts. To transmit a logical zero, the data+wire goes to zero volts, and the data-wire to positive five volts. This is quite different from the more common RS-232 interface, where one wire is always kept at zero volts. In RS-232, a logical one is transmitted by putting between positive six and positive twelve volts on the line, and a logical zero is transmitted by putting a voltage between negative six and negative twelve volts onto the line. RS-485 is generally understood to be better for data transmission than RS-232. With RS-232, the receiver has to measure if the incoming voltage is positive or negative. With RS-485, the receiver only needs to determine which line has the higher voltage on it.
The two wires over which RS-485 is transmitted are preferably twisted. Twisting means that disturbances on the line tend to affect both lines simultaneously, more or less by the same amount, so that the voltage on both lines will fluctuate, but the difference in voltage between the lines remains the same. The result is that noise is rejected from the line. Also, the drive capability of RS-485 drivers is higher than RS-232 drivers. As a result, the RS-485 protocol can connect devices over distances hundreds of times further than would be possible when using RS-232. RS-485 also increases the maximum data rate, i.e., the maximum amount of data which can be transmitted over the line every second. Communication between devices using RS-232 is normally about nine thousand six hundred baud (bits per second). Faster communication is possible, but the distances over which data can be transmitted are reduced significantly if communication is faster. By comparison, DMX-512 (using RS-485) permits data to be sent at two hundred fifty thousand baud (two hundred fifty thousand bits per second) over distances of hundreds of meters without problems. Every byte transmitted has one start bit, which is used to warn the receiver that the next character is starting, eight data bits (this conveys up to two hundred fifty six different levels) and two stop bits, which are used to tell the receiver that this is the end of the character. This means that every byte is transmitted as eleven bits, so that the length of each character is forty-four micro seconds.
The receiver looks at the two incoming signals on a pair of pins and compares the differences. A voltage rise on one wire and the inverse on the other will be seen as a differential and therefore deciphered as a digit. When both signals are identical, no difference is recognized and no digit deciphered. If interference was accidently transmitted along the line, it would impart no response as long as the interference was identical on both lines. The proximity of the two lines assist in assuring that distribution of interference is identical on both wires. The signal driver sends five hundred twelve device codes in a continual, repetitive stream of data. The receiving device is addressed with a number between one and five hundred twelve so it will respond only to data that corresponds to its assigned address.
A terminator resistor is typically installed at the end of a DMX line of devices, which reduces the possibility of signal reflection which can create errors in the DMX signal. The ohm value of the resistor is determined by the cable type used. Some devices allow for self termination at the end of the line. Multiple lines of DMX data can be distributed through an opto-repeater. This device creates a physical break in the line by transforming the electrical signals into light which spans a gap, then it is restored to electrical signals. This protects devices from damaging high voltage, accidentally traveling along the network. It will also repeat the original DMX data to several output lines. The input data is recreated at the outputs, eliminating distortion. The signal leaves the opto-repeater as strong as it left the console.
DMX messages are typically generated through computer software. Each DMX message is preceded with a "break," which is a signal for the receiver that the previous message has ended and the next message is about to start. The length of the break signal (equivalent to a logical zero on the line) has to be eighty-eight micro seconds according to the DMX specification. The signal can be more than eighty-eight micro seconds. After the break signal is removed from the line, there is a period during which the signal is at a logical one level. This is known as the "Mark" or `Mark After Break` (MAB) time. This time is typically at least eight micro seconds. After the Mark comes the first character, or byte, which is knows as the "Start" character. This character is rather loosely specified, and is normally set to the value zero (it can vary between zero and two hundred fifty five). This start character may be used to specify special messages. It is, for example, possible to have five hundred twelve dimmers which respond to messages with the start character set to zero, and another five hundred twelve dimmers which respond to messages with the start character set to one. If one transmits data for these one thousand twenty-four dimmers, and one sets the start character to zero for the first five hundred twelve dimmers, and to one for the second set of five hundred twelve dimmers, it is possible to control one thousand twenty four dimmers (or more if one wishes, using the same technique). The disadvantage is a reduction in the number of messages sent to each of the set of dimmers, in this example by a factor two. After the start character there are between one and five hundred twelve characters, which normally correspond to the up to five hundred twelve channels controlled by DMX. Each of these characters may have a value between zero (for `off`, zero percent) and two hundred fifty five (for full, one hundred percent). After the last character there may be another delay (at logic one level) before the next break starts. The number of messages which are transmitted every second are dependent on all the parameters listed above. In one case, where the break length is eighty-eight microseconds, the make after break length is eight micro seconds, and each character takes exactly forty-four micro seconds to transmit there will be forty-four messages per second, assuming that all five hundred twelve channels are being transmitted. Many lighting desks and other DMX sources transmit less than five hundred twelve channels, use a longer break and make after break time, and may have a refresh rate of seventy or eighty messages per second. Often, there is no benefit to be had from this, as the current value is not necessarily recalculated for each of the channels in each frame. The `standard` DMX signal would allow for a lamp to be switched on and off twenty-two times per second, which is ample for many applications. Certain devices are capable of using sixteen-bit DMX. Normal eight bit messages allow two hundred fifty-six positions, which is inadequate for the positioning of mirrors and other mechanical devices. Having sixteen bits available per channel increases that quantity up to sixty-five thousand five hundred thirty-six steps, which removes the limitation of `standard` DMX.
A significant problem with present lighting networks is that they require special wiring or cabling. In particular, one set of wires is needed for electrical power, while a second set of wires is needed for data, such as DMX-512 protocol data. Accordingly, the owner of an existing set of lights must undertake significant effort to rewire in order to have a digitally controlled lighting environment.
A second significant problem with present lighting networks is that particular lighting applications require particular lighting types. For example, LED based lights are appropriate for some applications, while incandescent lamps or halogen lamps may be more appropriate for other applications. A user who wishes to have a digitally controlled network of lights, in addition to rewiring, must currently add additional fixtures or replace old fixtures for each different type of light. Accordingly, a need has arisen for a lighting fixture that permits use of different types of digitally controlled lights.
Use of pulse width modulated signals to control electrical devices, such as motors, is also known. Traditional methods of providing pulse width modulated signals include hardware using software programmed timers, which in some instances is not cost effective if not enough timer modules are available, and one interrupt per count processes, in which a microprocessor receives periodic interrupts at a known rate. Each time through the interrupt loop the processor compares the current count with the target counts and updates one or more output pins, thus creating a pulse width modulated signal, or PWM. In this case, the speed equals the clock speed divided by cycles in the interrupt routine divided by desired resolution. In a third method, in a combination of the first two processes, software loops contain a variable number of instructions. The processor uses the hardware timer to generate a periodic interrupt, and then, depending on whether the pulse is to be very short or not, either schedules another interrupt to finish the PWM cycle, or creates the pulse by itself in the first interrupt routine by executing a series of instructions consuming a desired amount of time between two PWM signal updates. The difficulty with the third method is that for multiple PWM channels it is very difficult to arrange the timer based signal updates such that they do not overlap, and then to accurately change the update times for a new value of PWM signals. Accordingly, a new pulse width modulation method and system is needed to assisting in controlling electrical devices.
Many conventional illumination applications are subject to other drawbacks. Conventional light sources, such as halogen and incandescent sources may produce undesirable heat. Such sources may have very limited life spans. Conventional light sources may require substantial lens and filtering systems in order to produce color. It may be very difficult to reproduce precise color conditions with conventional light sources. Conventional light sources may not respond quickly to computer control. One or more of these drawbacks may have particular significance in particular existing lighting applications. Moreover, the combination of these drawbacks may have prevented the development of a number of other illumination applications. Accordingly, a need exists for illumination methods and systems that overcome the drawbacks of conventional illumination systems and that take advantage of the possibilities offered by overcoming such drawbacks.